Search Results for author: Tongliang Liu

Found 127 papers, 34 papers with code

LTF: A Label Transformation Framework for Correcting Label Shift

no code implementations ICML 2020 Jiaxian Guo, Mingming Gong, Tongliang Liu, Kun Zhang, DaCheng Tao

Distribution shift is a major obstacle to the deployment of current deep learning models on real-world problems.

Label-Noise Robust Domain Adaptation

no code implementations ICML 2020 Xiyu Yu, Tongliang Liu, Mingming Gong, Kun Zhang, Kayhan Batmanghelich, DaCheng Tao

Domain adaptation aims to correct the classifiers when faced with distribution shift between source (training) and target (test) domains.

Denoising Domain Adaptation

Dual-Path Distillation: A Unified Framework to Improve Black-Box Attacks

no code implementations ICML 2020 Yonggang Zhang, Ya Li, Tongliang Liu, Xinmei Tian

To obtain sufficient knowledge for crafting adversarial examples, previous methods query the target model with inputs that are perturbed with different searching directions.

Pluralistic Image Completion with Probabilistic Mixture-of-Experts

no code implementations18 May 2022 Xiaobo Xia, Wenhao Yang, Jie Ren, Yewen Li, Yibing Zhan, Bo Han, Tongliang Liu

Second, the constraints for diversity are designed to be task-agnostic, which causes the constraints to not work well.

Graph Pooling for Graph Neural Networks: Progress, Challenges, and Opportunities

1 code implementation15 Apr 2022 Chuang Liu, Yibing Zhan, Chang Li, Bo Du, Jia Wu, Wenbin Hu, Tongliang Liu, DaCheng Tao

Graph neural networks have emerged as a leading architecture for many graph-level tasks such as graph classification and graph generation with a notable improvement.

Graph Classification Graph Generation

SimT: Handling Open-set Noise for Domain Adaptive Semantic Segmentation

1 code implementation29 Mar 2022 Xiaoqing Guo, Jie Liu, Tongliang Liu, Yixuan Yuan

By exploiting computational geometry analysis and properties of segmentation, we design three complementary regularizers, i. e. volume regularization, anchor guidance, convex guarantee, to approximate the true SimT.

Semantic Segmentation

Killing Two Birds with One Stone:Efficient and Robust Training of Face Recognition CNNs by Partial FC

1 code implementation28 Mar 2022 Xiang An, Jiankang Deng, Jia Guo, Ziyong Feng, Xuhan Zhu, Jing Yang, Tongliang Liu

In each iteration, positive class centers and a random subset of negative class centers are selected to compute the margin-based softmax loss.

 Ranked #1 on Face Verification on IJB-C (TAR @ FAR=1e-4 metric)

Face Recognition Face Verification

Selective-Supervised Contrastive Learning with Noisy Labels

1 code implementation8 Mar 2022 Shikun Li, Xiaobo Xia, Shiming Ge, Tongliang Liu

In the selection process, by measuring the agreement between learned representations and given labels, we first identify confident examples that are exploited to build confident pairs.

Contrastive Learning Learning with noisy labels +1

Trustable Co-label Learning from Multiple Noisy Annotators

1 code implementation8 Mar 2022 Shikun Li, Tongliang Liu, Jiyong Tan, Dan Zeng, Shiming Ge

This raises the following important question: how can we effectively use a small amount of trusted data to facilitate robust classifier learning from multiple annotators?

Understanding and Improving Graph Injection Attack by Promoting Unnoticeability

1 code implementation ICLR 2022 Yongqiang Chen, Han Yang, Yonggang Zhang, Kaili Ma, Tongliang Liu, Bo Han, James Cheng

Recently Graph Injection Attack (GIA) emerges as a practical attack scenario on Graph Neural Networks (GNNs), where the adversary can merely inject few malicious nodes instead of modifying existing nodes or edges, i. e., Graph Modification Attack (GMA).

Invariance Principle Meets Out-of-Distribution Generalization on Graphs

no code implementations11 Feb 2022 Yongqiang Chen, Yonggang Zhang, Han Yang, Kaili Ma, Binghui Xie, Tongliang Liu, Bo Han, James Cheng

Despite recent developments in using the invariance principle from causality to enable out-of-distribution (OOD) generalization on Euclidean data, e. g., images, studies on graph data are limited.

Out-of-Distribution Generalization

Federated Causal Discovery

1 code implementation7 Dec 2021 Erdun Gao, Junjia Chen, Li Shen, Tongliang Liu, Mingming Gong, Howard Bondell

}$ In this paper, with the additive noise model assumption of data, we take the first step in developing a gradient-based learning framework named DAG-Shared Federated Causal Discovery (DS-FCD), which can learn the causal graph without directly touching local data and naturally handle the data heterogeneity.

Causal Discovery

Transfer Learning in Conversational Analysis through Reusing Preprocessing Data as Supervisors

no code implementations2 Dec 2021 Joshua Yee Kim, Tongliang Liu, Kalina Yacef

Conversational analysis systems are trained using noisy human labels and often require heavy preprocessing during multi-modal feature extraction.

Feature Engineering Multi-Task Learning

Confident Anchor-Induced Multi-Source Free Domain Adaptation

1 code implementation NeurIPS 2021 Jiahua Dong, Zhen Fang, Anjin Liu, Gan Sun, Tongliang Liu

To address these challenges, we develop a novel Confident-Anchor-induced multi-source-free Domain Adaptation (CAiDA) model, which is a pioneer exploration of knowledge adaptation from multiple source domains to the unlabeled target domain without any source data, but with only pre-trained source models.

Unsupervised Domain Adaptation

CRIS: CLIP-Driven Referring Image Segmentation

no code implementations30 Nov 2021 Zhaoqing Wang, Yu Lu, Qiang Li, Xunqiang Tao, Yandong Guo, Mingming Gong, Tongliang Liu

In addition, we present text-to-pixel contrastive learning to explicitly enforce the text feature similar to the related pixel-level features and dissimilar to the irrelevances.

Contrastive Learning Referring Expression Segmentation +1

Meta Clustering Learning for Large-scale Unsupervised Person Re-identification

no code implementations19 Nov 2021 Xin Jin, Tianyu He, Zhiheng Yin, Xu Shen, Tongliang Liu, Xinchao Wang, Jianqiang Huang, Xian-Sheng Hua, Zhibo Chen

Unsupervised Person Re-identification (U-ReID) with pseudo labeling recently reaches a competitive performance compared to fully-supervised ReID methods based on modern clustering algorithms.

Unsupervised Person Re-Identification

Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotations

1 code implementation ICLR 2022 Jiaheng Wei, Zhaowei Zhu, Hao Cheng, Tongliang Liu, Gang Niu, Yang Liu

These observations require us to rethink the treatment of noisy labels, and we hope the availability of these two datasets would facilitate the development and evaluation of future learning with noisy label solutions.

Learning with noisy labels

Co-variance: Tackling Noisy Labels with Sample Selection by Emphasizing High-variance Examples

no code implementations29 Sep 2021 Xiaobo Xia, Bo Han, Yibing Zhan, Jun Yu, Mingming Gong, Chen Gong, Tongliang Liu

The sample selection approach is popular in learning with noisy labels, which tends to select potentially clean data out of noisy data for robust training.

Learning with noisy labels

PI-GNN: Towards Robust Semi-Supervised Node Classification against Noisy Labels

no code implementations29 Sep 2021 Xuefeng Du, Tian Bian, Yu Rong, Bo Han, Tongliang Liu, Tingyang Xu, Wenbing Huang, Junzhou Huang

Semi-supervised node classification on graphs is a fundamental problem in graph mining that uses a small set of labeled nodes and many unlabeled nodes for training, so that its performance is quite sensitive to the quality of the node labels.

Graph Mining Node Classification

Modeling Adversarial Noise for Adversarial Defense

no code implementations29 Sep 2021 Dawei Zhou, Nannan Wang, Bo Han, Tongliang Liu

Deep neural networks have been demonstrated to be vulnerable to adversarial noise, promoting the development of defense against adversarial attacks.

Adversarial Defense

$\alpha$-Weighted Federated Adversarial Training

no code implementations29 Sep 2021 Jianing Zhu, Jiangchao Yao, Tongliang Liu, Kunyang Jia, Jingren Zhou, Bo Han, Hongxia Yang

Federated Adversarial Training (FAT) helps us address the data privacy and governance issues, meanwhile maintains the model robustness to the adversarial attack.

Adversarial Attack Federated Learning

Understanding Generalized Label Smoothing when Learning with Noisy Labels

no code implementations29 Sep 2021 Jiaheng Wei, Hangyu Liu, Tongliang Liu, Gang Niu, Yang Liu

It was shown that LS serves as a regularizer for training data with hard labels and therefore improves the generalization of the model.

Learning with noisy labels

Meta Discovery: Learning to Discover Novel Classes given Very Limited Data

no code implementations ICLR 2022 Haoang Chi, Feng Liu, Wenjing Yang, Long Lan, Tongliang Liu, Bo Han, Gang Niu, Mingyuan Zhou, Masashi Sugiyama

In this paper, we demystify assumptions behind L2DNC and find that high-level semantic features should be shared among the seen and unseen classes.

Meta-Learning

Exploiting Class Activation Value for Partial-Label Learning

no code implementations ICLR 2022 Fei Zhang, Lei Feng, Bo Han, Tongliang Liu, Gang Niu, Tao Qin, Masashi Sugiyama

As the first contribution, we empirically show that the class activation map (CAM), a simple technique for discriminating the learning patterns of each class in images, is surprisingly better at making accurate predictions than the model itself on selecting the true label from candidate labels.

Multi-class Classification Partial Label Learning

Unleash the Potential of Adaptation Models via Dynamic Domain Labels

no code implementations29 Sep 2021 Xin Jin, Tianyu He, Xu Shen, Songhua Wu, Tongliang Liu, Xinchao Wang, Jianqiang Huang, Zhibo Chen, Xian-Sheng Hua

In this paper, we propose an embarrassing simple yet highly effective adversarial domain adaptation (ADA) method for effectively training models for alignment.

Domain Adaptation

Can Label-Noise Transition Matrix Help to Improve Sample Selection and Label Correction?

no code implementations29 Sep 2021 Yu Yao, Xuefeng Li, Tongliang Liu, Alan Blair, Mingming Gong, Bo Han, Gang Niu, Masashi Sugiyama

Existing methods for learning with noisy labels can be generally divided into two categories: (1) sample selection and label correction based on the memorization effect of neural networks; (2) loss correction with the transition matrix.

Learning with noisy labels

Robust Weight Perturbation for Adversarial Training

no code implementations29 Sep 2021 Chaojian Yu, Bo Han, Mingming Gong, Li Shen, Shiming Ge, Bo Du, Tongliang Liu

In this paper, we propose such a criterion, namely Loss Stationary Condition (LSC) for constrained perturbation.

Rethinking Class-Prior Estimation for Positive-Unlabeled Learning

no code implementations ICLR 2022 Yu Yao, Tongliang Liu, Bo Han, Mingming Gong, Gang Niu, Masashi Sugiyama, DaCheng Tao

Hitherto, the distributional-assumption-free CPE methods rely on a critical assumption that the support of the positive data distribution cannot be contained in the support of the negative data distribution.

Modeling Adversarial Noise for Adversarial Training

no code implementations21 Sep 2021 Dawei Zhou, Nannan Wang, Bo Han, Tongliang Liu

Deep neural networks have been demonstrated to be vulnerable to adversarial noise, promoting the development of defense against adversarial attacks.

Adversarial Defense

Instance-dependent Label-noise Learning under a Structural Causal Model

1 code implementation NeurIPS 2021 Yu Yao, Tongliang Liu, Mingming Gong, Bo Han, Gang Niu, Kun Zhang

In particular, we show that properly modeling the instances will contribute to the identifiability of the label noise transition matrix and thus lead to a better classifier.

Exploring Set Similarity for Dense Self-supervised Representation Learning

no code implementations19 Jul 2021 Zhaoqing Wang, Qiang Li, Guoxin Zhang, Pengfei Wan, Wen Zheng, Nannan Wang, Mingming Gong, Tongliang Liu

By considering the spatial correspondence, dense self-supervised representation learning has achieved superior performance on various dense prediction tasks.

Instance Segmentation Keypoint Detection +3

Kernel Mean Estimation by Marginalized Corrupted Distributions

no code implementations10 Jul 2021 Xiaobo Xia, Shuo Shan, Mingming Gong, Nannan Wang, Fei Gao, Haikun Wei, Tongliang Liu

Estimating the kernel mean in a reproducing kernel Hilbert space is a critical component in many kernel learning algorithms.

Revisiting Knowledge Distillation: An Inheritance and Exploration Framework

1 code implementation CVPR 2021 Zhen Huang, Xu Shen, Jun Xing, Tongliang Liu, Xinmei Tian, Houqiang Li, Bing Deng, Jianqiang Huang, Xian-Sheng Hua

The inheritance part is learned with a similarity loss to transfer the existing learned knowledge from the teacher model to the student model, while the exploration part is encouraged to learn representations different from the inherited ones with a dis-similarity loss.

Knowledge Distillation

Understanding and Improving Early Stopping for Learning with Noisy Labels

1 code implementation NeurIPS 2021 Yingbin Bai, Erkun Yang, Bo Han, Yanhua Yang, Jiatong Li, Yinian Mao, Gang Niu, Tongliang Liu

Instead of the early stopping, which trains a whole DNN all at once, we initially train former DNN layers by optimizing the DNN with a relatively large number of epochs.

Learning with noisy labels

Probabilistic Margins for Instance Reweighting in Adversarial Training

1 code implementation NeurIPS 2021 Qizhou Wang, Feng Liu, Bo Han, Tongliang Liu, Chen Gong, Gang Niu, Mingyuan Zhou, Masashi Sugiyama

Reweighting adversarial data during training has been recently shown to improve adversarial robustness, where data closer to the current decision boundaries are regarded as more critical and given larger weights.

Adversarial Robustness

PI-GNN: A Novel Perspective on Semi-Supervised Node Classification against Noisy Labels

no code implementations14 Jun 2021 Xuefeng Du, Tian Bian, Yu Rong, Bo Han, Tongliang Liu, Tingyang Xu, Wenbing Huang, Junzhou Huang

Semi-supervised node classification, as a fundamental problem in graph learning, leverages unlabeled nodes along with a small portion of labeled nodes for training.

Graph Learning Node Classification

Adversarial Robustness through the Lens of Causality

no code implementations ICLR 2022 Yonggang Zhang, Mingming Gong, Tongliang Liu, Gang Niu, Xinmei Tian, Bo Han, Bernhard Schölkopf, Kun Zhang

The spurious correlation implies that the adversarial distribution is constructed via making the statistical conditional association between style information and labels drastically different from that in natural distribution.

Adversarial Attack Adversarial Robustness

TOHAN: A One-step Approach towards Few-shot Hypothesis Adaptation

no code implementations NeurIPS 2021 Haoang Chi, Feng Liu, Wenjing Yang, Long Lan, Tongliang Liu, Bo Han, William K. Cheung, James T. Kwok

To this end, we propose a target orientated hypothesis adaptation network (TOHAN) to solve the FHA problem, where we generate highly-compatible unlabeled data (i. e., an intermediate domain) to help train a target-domain classifier.

Domain Adaptation

KRADA: Known-region-aware Domain Alignment for Open World Semantic Segmentation

no code implementations11 Jun 2021 Chenhong Zhou, Feng Liu, Chen Gong, Tongliang Liu, Bo Han, William Cheung

However, in an open world, the unlabeled test images probably contain unknown categories and have different distributions from the labeled images.

Semantic Segmentation

Improving White-box Robustness of Pre-processing Defenses via Joint Adversarial Training

no code implementations10 Jun 2021 Dawei Zhou, Nannan Wang, Xinbo Gao, Bo Han, Jun Yu, Xiaoyu Wang, Tongliang Liu

However, pre-processing methods may suffer from the robustness degradation effect, in which the defense reduces rather than improving the adversarial robustness of a target model in a white-box setting.

Adversarial Defense Adversarial Robustness

Towards Defending against Adversarial Examples via Attack-Invariant Features

no code implementations9 Jun 2021 Dawei Zhou, Tongliang Liu, Bo Han, Nannan Wang, Chunlei Peng, Xinbo Gao

However, given the continuously evolving attacks, models trained on seen types of adversarial examples generally cannot generalize well to unseen types of adversarial examples.

Adversarial Robustness

Reliable Adversarial Distillation with Unreliable Teachers

1 code implementation ICLR 2022 Jianing Zhu, Jiangchao Yao, Bo Han, Jingfeng Zhang, Tongliang Liu, Gang Niu, Jingren Zhou, Jianliang Xu, Hongxia Yang

However, when considering adversarial robustness, teachers may become unreliable and adversarial distillation may not work: teachers are pretrained on their own adversarial data, and it is too demanding to require that teachers are also good at every adversarial data queried by students.

Adversarial Robustness

Sample Selection with Uncertainty of Losses for Learning with Noisy Labels

no code implementations NeurIPS 2021 Xiaobo Xia, Tongliang Liu, Bo Han, Mingming Gong, Jun Yu, Gang Niu, Masashi Sugiyama

In this way, we also give large-loss but less selected data a try; then, we can better distinguish between the cases (a) and (b) by seeing if the losses effectively decrease with the uncertainty after the try.

Learning with noisy labels

Instance Correction for Learning with Open-set Noisy Labels

no code implementations1 Jun 2021 Xiaobo Xia, Tongliang Liu, Bo Han, Mingming Gong, Jun Yu, Gang Niu, Masashi Sugiyama

Lots of approaches, e. g., loss correction and label correction, cannot handle such open-set noisy labels well, since they need training data and test data to share the same label space, which does not hold for learning with open-set noisy labels.

NoiLIn: Do Noisy Labels Always Hurt Adversarial Training?

no code implementations31 May 2021 Jingfeng Zhang, Xilie Xu, Bo Han, Tongliang Liu, Gang Niu, Lizhen Cui, Masashi Sugiyama

Adversarial training (AT) based on minimax optimization is a popular learning style that enhances the model's adversarial robustness.

Adversarial Robustness

Estimating Instance-dependent Label-noise Transition Matrix using DNNs

no code implementations27 May 2021 Shuo Yang, Erkun Yang, Bo Han, Yang Liu, Min Xu, Gang Niu, Tongliang Liu

Traditionally, the transition from clean distribution to noisy distribution (i. e., clean label transition matrix) has been widely exploited to learn a clean label classifier by employing the noisy data.

Relational Subsets Knowledge Distillation for Long-tailed Retinal Diseases Recognition

no code implementations22 Apr 2021 Lie Ju, Xin Wang, Lin Wang, Tongliang Liu, Xin Zhao, Tom Drummond, Dwarikanath Mahapatra, ZongYuan Ge

For example, there are estimated more than 40 different kinds of retinal diseases with variable morbidity, however with more than 30+ conditions are very rare from the global patient cohorts, which results in a typical long-tailed learning problem for deep learning-based screening models.

Knowledge Distillation

Removing Adversarial Noise in Class Activation Feature Space

no code implementations ICCV 2021 Dawei Zhou, Nannan Wang, Chunlei Peng, Xinbo Gao, Xiaoyu Wang, Jun Yu, Tongliang Liu

Then, we train a denoising model to minimize the distances between the adversarial examples and the natural examples in the class activation feature space.

Adversarial Robustness Denoising

Learning with Group Noise

no code implementations17 Mar 2021 Qizhou Wang, Jiangchao Yao, Chen Gong, Tongliang Liu, Mingming Gong, Hongxia Yang, Bo Han

Most of the previous approaches in this area focus on the pairwise relation (casual or correlational relationship) with noise, such as learning with noisy labels.

Learning with noisy labels

A Machine Learning Approach for Predicting Human Preference for Graph Layouts

no code implementations1 Mar 2021 Shijun Cai, Seok-Hee Hong, Jialiang Shen, Tongliang Liu

In this paper, we present the first machine learning approach for predicting human preference for graph layouts.

Transfer Learning

Improving Medical Image Classification with Label Noise Using Dual-uncertainty Estimation

no code implementations28 Feb 2021 Lie Ju, Xin Wang, Lin Wang, Dwarikanath Mahapatra, Xin Zhao, Mehrtash Harandi, Tom Drummond, Tongliang Liu, ZongYuan Ge

In this paper, we systematically discuss and define the two common types of label noise in medical images - disagreement label noise from inconsistency expert opinions and single-target label noise from wrong diagnosis record.

General Classification Image Classification +1

Demystifying Assumptions in Learning to Discover Novel Classes

no code implementations8 Feb 2021 Haoang Chi, Feng Liu, Wenjing Yang, Long Lan, Tongliang Liu, Bo Han, Gang Niu, Mingyuan Zhou, Masashi Sugiyama

In learning to discover novel classes (L2DNC), we are given labeled data from seen classes and unlabeled data from unseen classes, and we train clustering models for the unseen classes.

Meta-Learning

Understanding the Interaction of Adversarial Training with Noisy Labels

no code implementations6 Feb 2021 Jianing Zhu, Jingfeng Zhang, Bo Han, Tongliang Liu, Gang Niu, Hongxia Yang, Mohan Kankanhalli, Masashi Sugiyama

A recent adversarial training (AT) study showed that the number of projected gradient descent (PGD) steps to successfully attack a point (i. e., find an adversarial example in its proximity) is an effective measure of the robustness of this point.

Provably End-to-end Label-Noise Learning without Anchor Points

1 code implementation4 Feb 2021 Xuefeng Li, Tongliang Liu, Bo Han, Gang Niu, Masashi Sugiyama

In label-noise learning, the transition matrix plays a key role in building statistically consistent classifiers.

Learning with noisy labels

Learning Diverse-Structured Networks for Adversarial Robustness

1 code implementation3 Feb 2021 Xuefeng Du, Jingfeng Zhang, Bo Han, Tongliang Liu, Yu Rong, Gang Niu, Junzhou Huang, Masashi Sugiyama

In adversarial training (AT), the main focus has been the objective and optimizer while the model has been less studied, so that the models being used are still those classic ones in standard training (ST).

Adversarial Robustness

Tackling Instance-Dependent Label Noise via a Universal Probabilistic Model

1 code implementation14 Jan 2021 Qizhou Wang, Bo Han, Tongliang Liu, Gang Niu, Jian Yang, Chen Gong

The drastic increase of data quantity often brings the severe decrease of data quality, such as incorrect label annotations, which poses a great challenge for robustly training Deep Neural Networks (DNNs).

Score-based Causal Discovery from Heterogeneous Data

no code implementations1 Jan 2021 Chenwei Ding, Biwei Huang, Mingming Gong, Kun Zhang, Tongliang Liu, DaCheng Tao

Most algorithms in causal discovery consider a single domain with a fixed distribution.

Causal Discovery

ADD-Defense: Towards Defending Widespread Adversarial Examples via Perturbation-Invariant Representation

no code implementations1 Jan 2021 Dawei Zhou, Tongliang Liu, Bo Han, Nannan Wang, Xinbo Gao

Motivated by this observation, we propose a defense framework ADD-Defense, which extracts the invariant information called \textit{perturbation-invariant representation} (PIR) to defend against widespread adversarial examples.

ME-MOMENTUM: EXTRACTING HARD CONFIDENT EXAMPLES FROM NOISILY LABELED DATA

no code implementations ICCV 2021 Yingbin Bai, Tongliang Liu

To extract hard confident examples that contain non-simple patterns and are entangled with the inaccurately labeled examples, we borrow the idea of momentum from physics.

Learning with noisy labels

Improving robustness of softmax corss-entropy loss via inference information

no code implementations1 Jan 2021 Bingbing Song, wei he, Renyang Liu, Shui Yu, Ruxin Wang, Mingming Gong, Tongliang Liu, Wei Zhou

Several state-of-the-arts start from improving the inter-class separability of training samples by modifying loss functions, where we argue that the adversarial samples are ignored and thus limited robustness to adversarial attacks is resulted.

A Second-Order Approach to Learning with Instance-Dependent Label Noise

1 code implementation CVPR 2021 Zhaowei Zhu, Tongliang Liu, Yang Liu

We first provide evidences that the heterogeneous instance-dependent label noise is effectively down-weighting the examples with higher noise rates in a non-uniform way and thus causes imbalances, rendering the strategy of directly applying methods for class-dependent label noise questionable.

Image Classification Image Classification with Label Noise

COVID-MTL: Multitask Learning with Shift3D and Random-weighted Loss for Automated Diagnosis and Severity Assessment of COVID-19

no code implementations10 Dec 2020 Guoqing Bao, Huai Chen, Tongliang Liu, Guanzhong Gong, Yong Yin, Lisheng Wang, Xiuying Wang

In this paper, we present an end-to-end multitask learning (MTL) framework (COVID-MTL) that is capable of automated and simultaneous detection (against both radiology and NAT) and severity assessment of COVID-19.

COVID-19 Diagnosis Transfer Learning

Extended T: Learning with Mixed Closed-set and Open-set Noisy Labels

no code implementations2 Dec 2020 Xiaobo Xia, Tongliang Liu, Bo Han, Nannan Wang, Jiankang Deng, Jiatong Li, Yinian Mao

The traditional transition matrix is limited to model closed-set label noise, where noisy training data has true class labels within the noisy label set.

Domain Generalization via Entropy Regularization

1 code implementation NeurIPS 2020 Shanshan Zhao, Mingming Gong, Tongliang Liu, Huan Fu, DaCheng Tao

To arrive at this, some methods introduce a domain discriminator through adversarial learning to match the feature distributions in multiple source domains.

Domain Generalization

A Survey of Label-noise Representation Learning: Past, Present and Future

1 code implementation9 Nov 2020 Bo Han, Quanming Yao, Tongliang Liu, Gang Niu, Ivor W. Tsang, James T. Kwok, Masashi Sugiyama

Classical machine learning implicitly assumes that labels of the training data are sampled from a clean distribution, which can be too restrictive for real-world scenarios.

Learning Theory Representation Learning

Maximum Mean Discrepancy Test is Aware of Adversarial Attacks

2 code implementations22 Oct 2020 Ruize Gao, Feng Liu, Jingfeng Zhang, Bo Han, Tongliang Liu, Gang Niu, Masashi Sugiyama

However, it has been shown that the MMD test is unaware of adversarial attacks -- the MMD test failed to detect the discrepancy between natural and adversarial data.

Adversarial Attack Detection

Class2Simi: A New Perspective on Learning with Label Noise

no code implementations28 Sep 2020 Songhua Wu, Xiaobo Xia, Tongliang Liu, Bo Han, Mingming Gong, Nannan Wang, Haifeng Liu, Gang Niu

It is worthwhile to perform the transformation: We prove that the noise rate for the noisy similarity labels is lower than that of the noisy class labels, because similarity labels themselves are robust to noise.

Quantum differentially private sparse regression learning

no code implementations23 Jul 2020 Yuxuan Du, Min-Hsiu Hsieh, Tongliang Liu, Shan You, DaCheng Tao

The main contribution of this paper is devising a quantum DP Lasso estimator to earn the runtime speedup with the privacy preservation, i. e., the runtime complexity is $\tilde{O}(N^{3/2}\sqrt{d})$ with a nearly optimal utility bound $\tilde{O}(1/N^{2/3})$, where $N$ is the sample size and $d$ is the data dimension with $N\ll d$.

Part-dependent Label Noise: Towards Instance-dependent Label Noise

1 code implementation NeurIPS 2020 Xiaobo Xia, Tongliang Liu, Bo Han, Nannan Wang, Mingming Gong, Haifeng Liu, Gang Niu, DaCheng Tao, Masashi Sugiyama

Learning with the \textit{instance-dependent} label noise is challenging, because it is hard to model such real-world noise.

Dual T: Reducing Estimation Error for Transition Matrix in Label-noise Learning

1 code implementation NeurIPS 2020 Yu Yao, Tongliang Liu, Bo Han, Mingming Gong, Jiankang Deng, Gang Niu, Masashi Sugiyama

By this intermediate class, the original transition matrix can then be factorized into the product of two easy-to-estimate transition matrices.

Class2Simi: A Noise Reduction Perspective on Learning with Noisy Labels

no code implementations14 Jun 2020 Songhua Wu, Xiaobo Xia, Tongliang Liu, Bo Han, Mingming Gong, Nannan Wang, Haifeng Liu, Gang Niu

To give an affirmative answer, in this paper, we propose a framework called Class2Simi: it transforms data points with noisy class labels to data pairs with noisy similarity labels, where a similarity label denotes whether a pair shares the class label or not.

Contrastive Learning Learning with noisy labels +1

Repulsive Mixture Models of Exponential Family PCA for Clustering

no code implementations7 Apr 2020 Maoying Qiao, Tongliang Liu, Jun Yu, Wei Bian, DaCheng Tao

To alleviate this problem, in this paper, a repulsiveness-encouraging prior is introduced among mixing components and a diversified EPCA mixture (DEPCAM) model is developed in the Bayesian framework.

Quantum noise protects quantum classifiers against adversaries

no code implementations20 Mar 2020 Yuxuan Du, Min-Hsiu Hsieh, Tongliang Liu, DaCheng Tao, Nana Liu

This robustness property is intimately connected with an important security concept called differential privacy which can be extended to quantum differential privacy.

Classification General Classification

Multi-Class Classification from Noisy-Similarity-Labeled Data

no code implementations16 Feb 2020 Songhua Wu, Xiaobo Xia, Tongliang Liu, Bo Han, Mingming Gong, Nannan Wang, Haifeng Liu, Gang Niu

We further estimate the transition matrix from only noisy data and build a novel learning system to learn a classifier which can assign noise-free class labels for instances.

Classification General Classification +1

Towards Mixture Proportion Estimation without Irreducibility

no code implementations10 Feb 2020 Yu Yao, Tongliang Liu, Bo Han, Mingming Gong, Gang Niu, Masashi Sugiyama, DaCheng Tao

It is worthwhile to change the problem: we prove that if the assumption holds, our method will not affect anything; if the assumption does not hold, the bias from problem changing is less than the bias from violation of the irreducible assumption in the original problem.

Confidence Scores Make Instance-dependent Label-noise Learning Possible

no code implementations11 Jan 2020 Antonin Berthon, Bo Han, Gang Niu, Tongliang Liu, Masashi Sugiyama

We find with the help of confidence scores, the transition distribution of each instance can be approximately estimated.

Learning with noisy labels

A Shape Transformation-based Dataset Augmentation Framework for Pedestrian Detection

no code implementations15 Dec 2019 Zhe Chen, Wanli Ouyang, Tongliang Liu, DaCheng Tao

Alternatively, to access much more natural-looking pedestrians, we propose to augment pedestrian detection datasets by transforming real pedestrians from the same dataset into different shapes.

Pedestrian Detection

Control Batch Size and Learning Rate to Generalize Well: Theoretical and Empirical Evidence

no code implementations NeurIPS 2019 Fengxiang He, Tongliang Liu, DaCheng Tao

Specifically, we prove a PAC-Bayes generalization bound for neural networks trained by SGD, which has a positive correlation with the ratio of batch size to learning rate.

Continuous Dropout

1 code implementation28 Nov 2019 Xu Shen, Xinmei Tian, Tongliang Liu, Fang Xu, DaCheng Tao

On the one hand, continuous dropout is considerably closer to the activation characteristics of neurons in the human brain than traditional binary dropout.

Where is the Bottleneck of Adversarial Learning with Unlabeled Data?

no code implementations20 Nov 2019 Jingfeng Zhang, Bo Han, Gang Niu, Tongliang Liu, Masashi Sugiyama

Deep neural networks (DNNs) are incredibly brittle due to adversarial examples.

Towards Digital Retina in Smart Cities: A Model Generation, Utilization and Communication Paradigm

1 code implementation31 Jul 2019 Yihang Lou, Ling-Yu Duan, Yong Luo, Ziqian Chen, Tongliang Liu, Shiqi Wang, Wen Gao

The digital retina in smart cities is to select what the City Eye tells the City Brain, and convert the acquired visual data from front-end visual sensors to features in an intelligent sensing manner.

A Quantum-inspired Algorithm for General Minimum Conical Hull Problems

no code implementations16 Jul 2019 Yuxuan Du, Min-Hsiu Hsieh, Tongliang Liu, DaCheng Tao

In this paper, we propose a sublinear classical algorithm to tackle general minimum conical hull problems when the input has stored in a sample-based low-overhead data structure.

Truncated Cauchy Non-negative Matrix Factorization

no code implementations2 Jun 2019 Naiyang Guan, Tongliang Liu, Yangmuzi Zhang, DaCheng Tao, Larry S. Davis

Non-negative matrix factorization (NMF) minimizes the Euclidean distance between the data matrix and its low rank approximation, and it fails when applied to corrupted data because the loss function is sensitive to outliers.

Image Clustering

Are Anchor Points Really Indispensable in Label-Noise Learning?

1 code implementation NeurIPS 2019 Xiaobo Xia, Tongliang Liu, Nannan Wang, Bo Han, Chen Gong, Gang Niu, Masashi Sugiyama

Existing theories have shown that the transition matrix can be learned by exploiting \textit{anchor points} (i. e., data points that belong to a specific class almost surely).

Learning with noisy labels

Orthogonal Deep Neural Networks

1 code implementation15 May 2019 Kui Jia, Shuai Li, Yuxin Wen, Tongliang Liu, DaCheng Tao

To this end, we first prove that DNNs are of local isometry on data distributions of practical interest; by using a new covering of the sample space and introducing the local isometry property of DNNs into generalization analysis, we establish a new generalization error bound that is both scale- and range-sensitive to singular value spectrum of each of networks' weight matrices.

Image Classification

DistillHash: Unsupervised Deep Hashing by Distilling Data Pairs

no code implementations CVPR 2019 Erkun Yang, Tongliang Liu, Cheng Deng, Wei Liu, DaCheng Tao

To address this issue, we propose a novel deep unsupervised hashing model, dubbed DistillHash, which can learn a distilled data set consisted of data pairs, which have confidence similarity signals.

Semantic Similarity Semantic Textual Similarity

dipIQ: Blind Image Quality Assessment by Learning-to-Rank Discriminable Image Pairs

no code implementations13 Apr 2019 Kede Ma, Wentao Liu, Tongliang Liu, Zhou Wang, DaCheng Tao

One of the biggest challenges in learning BIQA models is the conflict between the gigantic image space (which is in the dimension of the number of image pixels) and the extremely limited reliable ground truth data for training.

Blind Image Quality Assessment Learning-To-Rank

Transferring Knowledge Fragments for Learning Distance Metric from A Heterogeneous Domain

no code implementations8 Apr 2019 Yong Luo, Yonggang Wen, Tongliang Liu, DaCheng Tao

Some existing heterogeneous transfer learning (HTL) approaches can learn target distance metric by usually transforming the samples of source and target domain into a common subspace.

Metric Learning Transfer Learning

Adaptive Morphological Reconstruction for Seeded Image Segmentation

1 code implementation8 Apr 2019 Tao Lei, Xiaohong Jia, Tongliang Liu, Shigang Liu, Hongy-ing Meng, Asoke K. Nandi

However, MR might mistakenly filter meaningful seeds that are required for generating accurate segmentation and it is also sensitive to the scale because a single-scale structuring element is employed.

Semantic Segmentation

Decomposition-Based Transfer Distance Metric Learning for Image Classification

no code implementations8 Apr 2019 Yong Luo, Tongliang Liu, DaCheng Tao, Chao Xu

In particular, DTDML learns a sparse combination of the base metrics to construct the target metric by forcing the target metric to be close to an integration of the source metrics.

Classification General Classification +3

Multi-View Matrix Completion for Multi-Label Image Classification

no code implementations8 Apr 2019 Yong Luo, Tongliang Liu, DaCheng Tao, Chao Xu

Therefore, we propose to weightedly combine the MC outputs of different views, and present the multi-view matrix completion (MVMC) framework for transductive multi-label image classification.

Classification General Classification +4

Fast Supervised Discrete Hashing

no code implementations7 Apr 2019 Jie Gui, Tongliang Liu, Zhenan Sun, DaCheng Tao, Tieniu Tan

Rather than adopting this method, FSDH uses a very simple yet effective regression of the class labels of training examples to the corresponding hash code to accelerate the algorithm.

A Regularization Approach for Instance-Based Superset Label Learning

no code implementations5 Apr 2019 Chen Gong, Tongliang Liu, Yuanyan Tang, Jian Yang, Jie Yang, DaCheng Tao

As a result, the intrinsic constraints among different candidate labels are deployed, and the disambiguated labels generated by RegISL are more discriminative and accurate than those output by existing instance-based algorithms.

On Better Exploring and Exploiting Task Relationships in Multi-Task Learning: Joint Model and Feature Learning

no code implementations3 Apr 2019 Ya Li, Xinmei Tian, Tongliang Liu, DaCheng Tao

The objective of our proposed method is to transform the features from different tasks into a common feature space in which the tasks are closely related and the shared parameters can be better optimized.

Multi-Task Learning

Why ResNet Works? Residuals Generalize

no code implementations2 Apr 2019 Fengxiang He, Tongliang Liu, DaCheng Tao

This paper studies the influence of residual connections on the hypothesis complexity of the neural network in terms of the covering number of its hypothesis space.

Generative-Discriminative Complementary Learning

no code implementations2 Apr 2019 Yanwu Xu, Mingming Gong, Junxiang Chen, Tongliang Liu, Kun Zhang, Kayhan Batmanghelich

The success of such approaches heavily depends on high-quality labeled instances, which are not easy to obtain, especially as the number of candidate classes increases.

Robust Angular Local Descriptor Learning

1 code implementation21 Jan 2019 Yanwu Xu, Mingming Gong, Tongliang Liu, Kayhan Batmanghelich, Chaohui Wang

In recent years, the learned local descriptors have outperformed handcrafted ones by a large margin, due to the powerful deep convolutional neural network architectures such as L2-Net [1] and triplet based metric learning [2].

Metric Learning

An Optimal Transport View on Generalization

no code implementations8 Nov 2018 Jingwei Zhang, Tongliang Liu, DaCheng Tao

We derive upper bounds on the generalization error of learning algorithms based on their \emph{algorithmic transport cost}: the expected Wasserstein distance between the output hypothesis and the output hypothesis conditioned on an input example.

Learning Theory

The Expressive Power of Parameterized Quantum Circuits

no code implementations29 Oct 2018 Yuxuan Du, Min-Hsiu Hsieh, Tongliang Liu, DaCheng Tao

Parameterized quantum circuits (PQCs) have been broadly used as a hybrid quantum-classical machine learning scheme to accomplish generative tasks.

Tensor Networks

Implementable Quantum Classifier for Nonlinear Data

no code implementations17 Sep 2018 Yuxuan Du, Min-Hsiu Hsieh, Tongliang Liu, DaCheng Tao

Ultimately, a stronger nonlinear classifier can be established, the so-called quantum ensemble learning (QEL), by combining a set of weak VQPs produced using a subsampling method.

Ensemble Learning

Deep Domain Generalization via Conditional Invariant Adversarial Networks

no code implementations ECCV 2018 Ya Li, Xinmei Tian, Mingming Gong, Yajing Liu, Tongliang Liu, Kun Zhang, DaCheng Tao

Under the assumption that the conditional distribution $P(Y|X)$ remains unchanged across domains, earlier approaches to domain generalization learned the invariant representation $T(X)$ by minimizing the discrepancy of the marginal distribution $P(T(X))$.

Domain Generalization Representation Learning

Correcting the Triplet Selection Bias for Triplet Loss

1 code implementation ECCV 2018 Baosheng Yu, Tongliang Liu, Mingming Gong, Changxing Ding, DaCheng Tao

Considering that the number of triplets grows cubically with the size of training data, triplet mining is thus indispensable for efficiently training with triplet loss.

Face Recognition Fine-Grained Image Classification +4

Instance-Dependent PU Learning by Bayesian Optimal Relabeling

no code implementations7 Aug 2018 Fengxiang He, Tongliang Liu, Geoffrey I. Webb, DaCheng Tao

Specifically, by treating the unlabelled data as noisy negative examples, we could automatically label a group positive and negative examples whose labels are identical to the ones assigned by a Bayesian optimal classifier with a consistency guarantee.

Domain Generalization via Conditional Invariant Representation

1 code implementation23 Jul 2018 Ya Li, Mingming Gong, Xinmei Tian, Tongliang Liu, DaCheng Tao

With the conditional invariant representation, the invariance of the joint distribution $\mathbb{P}(h(X), Y)$ can be guaranteed if the class prior $\mathbb{P}(Y)$ does not change across training and test domains.

Domain Generalization

An Efficient and Provable Approach for Mixture Proportion Estimation Using Linear Independence Assumption

no code implementations CVPR 2018 Xiyu Yu, Tongliang Liu, Mingming Gong, Kayhan Batmanghelich, DaCheng Tao

In this paper, we study the mixture proportion estimation (MPE) problem in a new setting: given samples from the mixture and the component distributions, we identify the proportions of the components in the mixture distribution.

Bayesian Quantum Circuit

no code implementations27 May 2018 Yuxuan Du, Tongliang Liu, DaCheng Tao

Parameterized quantum circuits (PQCs), as one of the most promising schemes to realize quantum machine learning algorithms on near-term quantum computers, have been designed to solve machine earning tasks with quantum advantages.

Quantum Physics

Semantic Structure-based Unsupervised Deep Hashing

1 code implementation IJCAI2018 2018 Erkun Yang, Cheng Deng, Tongliang Liu, Wei Liu, DaCheng Tao

Hashing is becoming increasingly popular for approximate nearest neighbor searching in massive databases due to its storage and search efficiency.

Semantic Similarity Semantic Textual Similarity

An Information-Theoretic View for Deep Learning

no code implementations24 Apr 2018 Jingwei Zhang, Tongliang Liu, DaCheng Tao

This upper bound shows that as the number of convolutional and pooling layers $L$ increases in the network, the expected generalization error will decrease exponentially to zero.

Speech Recognition

On the Rates of Convergence from Surrogate Risk Minimizers to the Bayes Optimal Classifier

no code implementations11 Feb 2018 Jingwei Zhang, Tongliang Liu, DaCheng Tao

We study the rates of convergence from empirical surrogate risk minimizers to the Bayes optimal classifier.

Learning with Biased Complementary Labels

1 code implementation ECCV 2018 Xiyu Yu, Tongliang Liu, Mingming Gong, DaCheng Tao

We therefore reason that the transition probabilities will be different.

Learning with Bounded Instance- and Label-dependent Label Noise

no code implementations ICML 2020 Jiacheng Cheng, Tongliang Liu, Kotagiri Ramamohanarao, DaCheng Tao

Inspired by the idea of learning with distilled examples, we then propose a learning algorithm with theoretical guarantees for its robustness to BILN.

Transfer Learning with Label Noise

no code implementations31 Jul 2017 Xiyu Yu, Tongliang Liu, Mingming Gong, Kun Zhang, Kayhan Batmanghelich, DaCheng Tao

However, when learning this invariant knowledge, existing methods assume that the labels in source domain are uncontaminated, while in reality, we often have access to source data with noisy labels.

Denoising Transfer Learning

On Compressing Deep Models by Low Rank and Sparse Decomposition

no code implementations CVPR 2017 Xiyu Yu, Tongliang Liu, Xinchao Wang, DaCheng Tao

Deep compression refers to removing the redundancy of parameters and feature maps for deep learning models.

Algorithmic stability and hypothesis complexity

no code implementations ICML 2017 Tongliang Liu, Gábor Lugosi, Gergely Neu, DaCheng Tao

The bounds are based on martingale inequalities in the Banach space to which the hypotheses belong.

Deep Blur Mapping: Exploiting High-Level Semantics by Deep Neural Networks

no code implementations5 Dec 2016 Kede Ma, Huan Fu, Tongliang Liu, Zhou Wang, DaCheng Tao

The human visual system excels at detecting local blur of visual images, but the underlying mechanism is not well understood.

Elastic Net Hypergraph Learning for Image Clustering and Semi-supervised Classification

no code implementations3 Mar 2016 Qingshan Liu, Yubao Sun, Cantian Wang, Tongliang Liu, DaCheng Tao

In the second step, hypergraph is used to represent the high order relationships between each datum and its prominent samples by regarding them as a hyperedge.

General Classification graph construction +2

Dimensionality-Dependent Generalization Bounds for $k$-Dimensional Coding Schemes

no code implementations3 Jan 2016 Tongliang Liu, DaCheng Tao, Dong Xu

Can we obtain dimensionality-dependent generalization bounds for $k$-dimensional coding schemes that are tighter than dimensionality-independent bounds when data is in a finite-dimensional feature space?

Dictionary Learning Generalization Bounds +1

Local Rademacher Complexity for Multi-label Learning

no code implementations26 Oct 2014 Chang Xu, Tongliang Liu, DaCheng Tao, Chao Xu

We analyze the local Rademacher complexity of empirical risk minimization (ERM)-based multi-label learning algorithms, and in doing so propose a new algorithm for multi-label learning.

Multi-Label Learning

Cannot find the paper you are looking for? You can Submit a new open access paper.